InterconnectsArtifacts Hub
withVAILVAIL
All models

DeepSeek-OCR

byDeepSeekDeepSeek· 17 Oct 2025
MultimodalSpecial purpose

A rather surprising release by DeepSeek showcasing the superior compression ratio of images to text as the main argument in their paper.

Specs
Params3.3B
LicenseMIT
Tags
modelsmodels/llms/multi-modalmultilingualartifacts/16
Resources
Hugging Face — deepseek-ai/DeepSeek-OCRPaper / tech reportAppeared in Artifacts Log #16
Adoption · Interconnects
RAM score
—
Hugging Face Downloads
2.5M
last 30d
29.2M
all time
HF Likes
3.3K

Relative Adoption Metric contextualizes downloads against the model's size bucket.

Related Models

More from DeepSeek

DeepSeek-V4-Flash
DeepSeekDeepSeek24 Apr 2026
DeepSeek-OCR-2
DeepSeekDeepSeek27 Jan 2026
DeepSeek-V3.2
DeepSeekDeepSeek01 Dec 2025
DeepSeek-V3.2-Exp
DeepSeekDeepSeek29 Sep 2025

Similar Models

Unlimited-OCR
Baidu (ERNIE)Baidu (ERNIE)22 Jun 2026
dots.mocr
RedNoteRedNote19 Mar 2026
Nanonets-OCR2-3B
NanonetsNanonets13 Oct 2025
PaliGemma 2 3B Mix 448
GoogleGoogle21 Nov 2024